[IA64] On return from syscall, hint b7 with __kernel_syscall_via_epc().

Why is this a good idea?  Clearing b7 to 0 is guaranteed to do us no
good and writing it with __kernel_syscall_via_epc() yields a 6 cycle
improvement _if_ the application performs another EPC-based system-
call without overwriting b7, which is not all that uncommon.  Well
worth the minimal cost of 1 bundle of code.

Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>

diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index 0c84bed1bda9e5090302782b92bfad629f76bdb6..6359d7ffbb7cef98474ccef247940ee3df477d1c 100644 (file)
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -692,7 +692,7 @@ ENTRY(ia64_leave_syscall)
        ;;
 (p6)   ld4 r31=[r18]                           // load current_thread_info()->flags
        ld8 r19=[r2],PT(B6)-PT(LOADRS)          // load ar.rsc value for "loadrs"
-       mov b7=r0               // clear b7
+       nop.i 0
        ;;
        ld8 r23=[r3],PT(R11)-PT(AR_BSPSTORE)    // load ar.bspstore (may be garbage)
        ld8 r18=[r2],PT(R9)-PT(B6)              // load b6
@@ -754,7 +754,14 @@ ENTRY(ia64_leave_syscall)
 (pUStk)        ld4 r17=[r3]            // r17 = cpu_data->phys_stacked_size_p8
        mov.m ar.csd=r0         // M2 clear ar.csd
        mov f10=f0              // clear f10
+
+       nop.m 0
+       movl r14=__kernel_syscall_via_epc       // X
        ;;
+       nop.m 0
+       nop.m 0
+       mov b7=r14              // I0   clear b7 (hint with __kernel_syscall_via_epc)
+
        mov r14=r0              // clear r14
        mov f11=f0              // clear f11
 (pKStk) br.cond.dpnt.many skip_rbs_switch